Field Planning: Fewer Kilometres, More Interventions Completed
A regional energy distributor was planning its 25 field technicians on spreadsheets, with overlapping rounds and lost hours. In a few months, we brought order to the planning process, using the tools already in place.
The company
A regional energy distributor whose 25 field technicians work on the network and at customer sites, with planning managed on spreadsheets.
The problem and its cost
Planning relied on several spreadsheets updated by hand by the dispatch team. Each technician received a list the day before, with no real logic by zone or skill. Neighbouring jobs were given to different people, parts were often missing from the vehicle, and every urgent call meant reshuffling everything.
This cost stayed invisible until someone calculated it. We estimated it at 495,000 CHF per year: 25 × 40 km × 220 days × 0.75 + 25 × 0.8 h × 220 days × 75 CHF, which is 165,000 CHF of avoidable kilometres and 330,000 CHF of unproductive hours. It is not a budget line, it is a diffuse loss spread across everyone's working day.
The strategy
We decided to start with facts before any rule: rebuilding the history of interventions, their actual durations, locations and the reasons for returning to the depot. Without that base, the new organisation would have rested on impressions, and the dispatch team would not have trusted it.
Only then did we set the planning rules (zones, skills, parts, handling of urgent calls), deploy the daily routine in the existing tools, and finally put steering in place. This order avoids equipping an organisation that is not yet clear, and lets us measure the effect of each step.
The execution
Team: We mobilised a dedicated contact, the WALVIS core team (strategy, finance, operations, data, AI) and a sector expert profile in energy distribution network operations, drawn from our network.
- →A cleaned and structured intervention base: job types, actual durations, locations, required parts, reasons for returning to the depot.
- →A set of planning rules by zone and skill, with a clear procedure for urgent calls.
- →A weekly planning and daily readjustment model, built in the spreadsheets and tools the dispatch team already used.
- →A steering dashboard reviewed weekly: kilometres driven, travel time, jobs resolved on the first visit.
We used AI in only two places: estimating the likely duration of a job from historical data, and suggesting a visiting order within a zone. The dispatch team keeps the decision and corrects the suggestion whenever the field requires it.
Most of the gain comes from organisation: stable zones, vehicles loaded according to the next day's jobs, a ten-minute briefing each morning. These simple rules need no algorithm and last because the teams understand them.
The result
Expected annual saving: 30% of CHF 495,000, about CHF 149,000. For an investment of about CHF 55,000, payback comes in 4 months.
We assume 30% of the current cost, not 100%, because part of the kilometres and hours cannot be removed (urgent calls, network geography, unforeseen faults) and new practices take time to settle in.
- More predictable days for technicians, with fewer back-and-forth trips and less waiting.
- Appointments kept more reliably for customers and a faster response to urgent calls.
- A dispatch team that decides on shared data rather than in permanent firefighting mode.
The assumptions
- 25 field technicians, 220 working days per year.
- 40 avoidable kilometres per day per technician, valued at 0.75 CHF per kilometre.
- 0.8 unproductive hours per day per technician, valued at a loaded hourly cost of 75 CHF.
- Investment of about 55,000 CHF, giving a payback of about 4 months on the expected saving.
Scenario type, built from real missions and anonymized.
In the same cocoon
